Abstract

Atmospheric simulation data present richer information in terms of spatiotemporal resolution, spatial dimension, and the number of physical quantities compared to observational data; however, such simulations do not perfectly correspond to the real atmospheric conditions. Additionally, extensive simulation data aids machine learning-based image classification in atmospheric science. In this study, we applied a machine learning model for tropical cyclone detection, which was trained using both simulation and satellite observation data. Consequently, the classification performance was significantly lower than that obtained with the application of simulation data. Owing to the large gap between the simulation and observation data, the classification model could not be practically trained only on the simulation data. Thus, the representation capability of the simulation data must be analyzed and integrated into the observation data for application in real problems.

Highlights

  • Deep learning is a machine learning method that uses multilayered neural networks; recently, it has been used to detect objects and structures in the field of atmospheric science

  • If the simulated data can interpolate a small number of observed cases, it could contribute toward improving the performance of the machine learning-based models for recognizing extreme events

  • This paper reports the initial results of applying a classification model developed by training only simulation data to satellite observation data, considering typhoon classification as a simple example

Read more

Summary

Introduction

Deep learning is a machine learning method that uses multilayered neural networks; recently, it has been used to detect objects and structures in the field of atmospheric science. Deep convolutional neural networks (DCNNs) specialized for image pattern recognition have exhibited excellent performance in detecting and/or classifying tropical cyclones (TCs) (Matsuoka et al, 2018), cloud type (Gorooh et al, 2020), weather fronts (Biard & Kunkel, 2019), and atmospheric river (Prabhat et al, 2021) from atmospheric data. Numerical simulations employing an atmospheric model can generate data of extreme events under various initial conditions and scenarios, they do not perfectly correspond to the real atmospheric conditions. If the simulated data can interpolate a small number of observed cases, it could contribute toward improving the performance of the machine learning-based models for recognizing extreme events. This paper reports the initial results of applying a classification model developed by training only simulation data to satellite observation data, considering typhoon classification as a simple example

Results
Discussion
Conclusion

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.